Instructions to use smashingtags/nova-talker-2b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use smashingtags/nova-talker-2b with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="smashingtags/nova-talker-2b",
	filename="gemma-2-2b-it-Q4_K_M.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use smashingtags/nova-talker-2b with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf smashingtags/nova-talker-2b:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf smashingtags/nova-talker-2b:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf smashingtags/nova-talker-2b:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf smashingtags/nova-talker-2b:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf smashingtags/nova-talker-2b:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf smashingtags/nova-talker-2b:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf smashingtags/nova-talker-2b:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf smashingtags/nova-talker-2b:Q4_K_M

Use Docker

docker model run hf.co/smashingtags/nova-talker-2b:Q4_K_M

LM Studio
Jan

vLLM

How to use smashingtags/nova-talker-2b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "smashingtags/nova-talker-2b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "smashingtags/nova-talker-2b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/smashingtags/nova-talker-2b:Q4_K_M

Ollama
How to use smashingtags/nova-talker-2b with Ollama:
```
ollama run hf.co/smashingtags/nova-talker-2b:Q4_K_M
```

Unsloth Studio

How to use smashingtags/nova-talker-2b with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for smashingtags/nova-talker-2b to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for smashingtags/nova-talker-2b to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for smashingtags/nova-talker-2b to start chatting

Atomic Chat new
Docker Model Runner
How to use smashingtags/nova-talker-2b with Docker Model Runner:
```
docker model run hf.co/smashingtags/nova-talker-2b:Q4_K_M
```

Lemonade

How to use smashingtags/nova-talker-2b with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull smashingtags/nova-talker-2b:Q4_K_M

Run and chat with the model

lemonade run user.nova-talker-2b-Q4_K_M

List all available models

lemonade list

Nova Talker 2B

Nova Talker 2B is the default conversational voice for Eight.ly OS's built-in AI assistant, Nova. It is Google Gemma 2 2B (instruction-tuned), quantized to Q4_K_M (~1.6 GB), re-hosted here so an Eight.ly OS install pulls the whole Nova stack from a single source it controls — no dependency on the public Ollama registry.

Role in Nova

Nova is router-in-front: a single hidden tool router — Nova Router 1.5B (Qwen2.5-Coder 1.5B, 99% tool-pick accuracy) — decides which NAS action to run for every talker. The talker's only job is to write the natural-language reply from the tool's real result. Nova Talker 2B is the smallest, fastest talker and the default installed on first boot; it runs on any CPU or GPU.

Talker	Base	Size	Role
Nova Talker 2B (this repo)	Gemma 2 2B	1.6 GB	Default voice
Nova Talker 4B	Qwen 3 4B	2.4 GB	Optional, richer wording
Nova Talker 12B	Gemma 4 12B	7.0 GB	Premium, GPU-only, experimental

License

Apache-2.0. Gemma is distributed under Apache-2.0 (https://ai.google.dev/gemma/apache_2).

Downloads last month: 22

GGUF

Model size

3B params

Architecture

gemma2

Hardware compatibility

4-bit

Model tree for smashingtags/nova-talker-2b

Base model

google/gemma-2-2b

Finetuned

google/gemma-2-2b-it

Quantized

(187)

this model